摘要
一位新闻记者-机器人与机器学习每日新闻-神经退行性疾病和并发症的新研究-帕金森氏症是一篇报道的主题。根据来自印度Chhattisgarh的新闻报道，由NewsRx记者报道，研究表明：“帕金森氏病(PD)是一种进行性神经退行性疾病，影响运动和非运动症状。其症状发展缓慢，难以早期识别。”我们的新闻编辑从雷普尔国家理工学院的一项研究中获得了一句话:“机器学习在预测语音数据中隐藏的特征方面具有重要的潜力。这项工作旨在从高维数据集中识别出最相关的特征。”本文分析了3种具有不同医学特征的基于语音的个体da特征集,提出了一种基于滤波器、w rapper和提取高度相关特征的嵌入算法的集成特征选择算法(EFSA)。采用了k近邻(KNN)、随机森林、决策树、支持向量机(SVM)、Bagging分类器、多层感知器(MLP)分类器、多层感知器(MLP)分类器等不同的ML模型,并在三个不同的基于语音的数据集上进行了验证,这些方法可以缩短训练时间,提高模型的精度,减少过拟合。此外，除了这些已建立的分类器之外，我们还在最优多数选票上找到了一个集成分类器，DataSet-I的分类准确率为97.6%，F1-score为97.9%，准确率为98%，召回率为98%，DataSet-II的分类准确率为90.2%。F1-Score 90.2%,Precision 90.2%,Recall 90.5%。DataSet-III的准确率为83.3%,F1-Score 83.3%,Precision 83.5%,Recall 83.3%。从各自的数据集中提取了23个特征中的13个,754个特征中的45个,46个特征中的17个。
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News-New research on Neurodegenerative Diseases and Co nditions-Parkinson's Disease is the subject of a report. According to news rep orting originating from Chhattisgarh, India, by NewsRx correspondents, research stated, "Parkinson's Disease (PD) is a progressive neurodegenerative disorder af fecting motor and non-motor symptoms. Its symptoms develop slowly, making early identification difficult." Our news editors obtained a quote from the research from the National Institute of Technology Raipur, "Machine learning has a significant potential to predict P arkinson's disease on features hidden in voice data. This work aimed to identify the most relevant features from a high-dimensional dataset, which helps accurat ely classify Parkinson's Disease with less computation time. Three individual da tasets with various medical features based on voice have been analyzed in this w ork. An Ensemble Feature Selection Algorithm (EFSA) technique based on filter, w rapper, and embedding algorithms that pick highly relevant features for identify ing Parkinson's Disease is proposed, and the same has been validated on three di fferent datasets based on voice. These techniques can shorten training time to i mprove model accuracy and minimize overfitting. We utilized different ML models such as K-Nearest Neighbors (KNN), Random Forest, Decision Tree, Support Vector Machine (SVM), Bagging Classifier, Multi-Layer Perceptron (MLP) Classifier, and Gradient Boosting. Each of these models was fine-tuned to ensure optimal perform ance within our specific context. Moreover, in addition to these established cla ssifiers, we proposed an ensemble classifier is found on a high optimal majority of the votes. Dataset-I achieves classification accuracy with 97.6 % , F1-score 97.9 %, precision with 98 % and recall wit h 98 %. Dataset-II achieves classification accuracy 90.2 % , F1-score 90.2 %, precision 90.2 %, and recall 90.5 % . Dataset-III achieves 83.3 % accuracy, F1-score 83.3 % , precision 83.5 % and recall 83.3 %. These results h ave been taken using 13 out of 23, 45 out of 754, and 17 out of 46 features from respective datasets."
NETL
NSTL